Necrotizing enterocolitis (NEC), a hemorrhagic inflammatory necrosis of the distal ileum, is the most common gastrointestinal emergency of premature infants. While advances in neonatal practice have improved the survival of infants born prematurely, the incidence of NEC has not decreased. Regrettably, the pathophysiology of this disease remains poorly understood, treatment is mainly supportive and no predictive diagnostic tests are available. Using neonatal rat and mouse models of NEC, we were the first to show that bile acids (BAs) play a crucial role in the development of this disease. Our preliminary data show fecal BA levels in premature infants that develop NEC have an increased coefficient of variability compared with those that do not develop disease. Importantly, these differences are seen days prior to diagnosis. We hypothesize that variability in fecal BAs could be utilized to predict disease development at its earliest stages when less invasive medical interventions can be applied. The goals of this proposal are to determine if fecal BAs can be used as a biomarker to develop the first predictive test for NEC. Using feces collected prospectively from premature infants, these goals will be accomplished with the following specific aims: 1) Test the hypothesis that total fecal BAs are more variable in infants that develop NEC. 2) Test the hypothesis that fecal BAs of infants that develop NEC are comprised of the more cytotoxic, hydrophobic BAs. Completion of these aims will further elucidate the mechanisms of BAs in NEC pathogenesis, advance the paradigm to translational studies and could lead to the development of the first predictive test for this devastating disease.